Shankar R. Rao

Motion Segmentation in the Presence of Outlying, Incomplete, or Corrupted Trajectories

In this paper, we study the problem of segmenting tracked feature point trajectories of multiple moving objects in an image sequence. Using the affine camera model, this problem can be cast as the problem of segmenting samples drawn from multiple linear subspaces. In practice, due to limitations of the tracker, occlusions, and the presence of nonrigid objects in the scene, the obtained motion trajectories may contain grossly mistracked features, missing entries, or corrupted entries. In this paper, we develop a robust subspace separation scheme that deals with these practical issues in a unified mathematical framework. Our methods draw strong connections between lossy compression, rank minimization, and sparse representation. We test our methods extensively on the Hopkins155 motion segmentation database and other motion sequences with outliers and missing data. We compare the performance of our methods to state-of-the-art motion segmentation methods based on expectation-maximization and spectral clustering. For data without outliers or missing information, the results of our methods are on par with the state-of-the-art results and, in many cases, exceed them. In addition, our methods give surprisingly good performance in the presence of the three types of pathological trajectories mentioned above. All code and results are publicly available at http://perception.csl.uiuc.edu/coding/motion/.

Motion segmentation via robust subspace separation in the presence of outlying, incomplete, or corrupted trajectories

We examine the problem of segmenting tracked feature point trajectories of multiple moving objects in an image sequence. Using the affine camera model, this motion segmentation problem can be cast as the problem of segmenting samples drawn from a union of linear subspaces. Due to limitations of the tracker, occlusions and the presence of nonrigid objects in the scene, the obtained motion trajectories may contain grossly mistracked features, missing entries, or not correspond to any valid motion model. In this paper, we develop a robust subspace separation scheme that can deal with all of these practical issues in a unified framework. Our methods draw strong connections between lossy compression, rank minimization, and sparse representation. We test our methods extensively and compare their performance to several extant methods with experiments on the Hopkins 155 database. Our results are on par with state-of-the-art results, and in many cases exceed them. All MATLAB code and segmentation results are publicly available for peer evaluation at http://perception.csl.uiuc.edu/coding/motion/.

Segmentation of Natural Images by Texture and Boundary Compression

We present a novel algorithm for segmentation of natural images that harnesses the principle of minimum description length (MDL). Our method is based on observations that a homogeneously textured region of a natural image can be well modeled by a Gaussian distribution and the region boundary can be effectively coded by an adaptive chain code. The optimal segmentation of an image is the one that gives the shortest coding length for encoding all textures and boundaries in the image, and is obtained via an agglomerative clustering process applied to a hierarchy of decreasing window sizes as multi-scale texture features. The optimal segmentation also provides an accurate estimate of the overall coding length and hence the true entropy of the image. We test our algorithm on the publicly available Berkeley Segmentation Dataset. It achieves state-of-the-art segmentation results compared to other existing methods.

Natural image segmentation with adaptive texture and boundary encoding

We present a novel algorithm for unsupervised segmentation of natural images that harnesses the principle of minimum description length (MDL). Our method is based on observations that a homogeneously textured region of a natural image can be well modeled by a Gaussian distribution and the region boundary can be effectively coded by an adaptive chain code. The optimal segmentation of an image is the one that gives the shortest coding length for encoding all textures and boundaries in the image, and is obtained via an agglomerative clustering process applied to a hierarchy of decreasing window sizes. The optimal segmentation also provides an accurate estimate of the overall coding length and hence the true entropy of the image. Our algorithm achieves state-of-the-art results on the Berkeley Segmentation Dataset compared to other popular methods.

Robust Statistical Estimation and Segmentation of Multiple Subspaces

We study the problem of estimating a mixed geometric model of multiple subspaces in the presence of a significant amount of outliers. The estimation of multiple subspaces is an important problem in computer vision, particularly for segmenting multiple motions in an image sequence. We first provide a comprehensive survey of robust statistical techniques in the literature, and identify three main approaches for detecting and rejecting outliers. Through a careful examination of these approaches, we propose and investigate three principled methods for robustly estimating mixed subspace models: random sample consensus, the influence function, and multivariate trimming. Using a benchmark synthetic experiment and a set of real image sequences, we conduct a thorough comparison of the three methods

Robust Algebraic Segmentation of Mixed Rigid-Body and Planar Motions from Two Views

This paper studies segmentation of multiple rigid-body motions in a 3-D dynamic scene under perspective camera projection. We consider dynamic scenes that contain both 3-D rigid-body structures and 2-D planar structures. Based on the well-known epipolar and homography constraints between two views, we propose a hybrid perspective constraint (HPC) to unify the representation of rigid-body and planar motions. Given a mixture of K hybrid perspective constraints, we propose an algebraic process to partition image correspondences to the individual 3-D motions, called Robust Algebraic Segmentation (RAS). Particularly, we prove that the joint distribution of image correspondences is uniquely determined by a set of (2K)-th degree polynomials, a global signature for the union of K motions of possibly mixed type. The first and second derivatives of these polynomials provide a means to recover the association of the individual image samples to their respective motions. Finally, using robust statistics, we show that the polynomials can be robustly estimated in the presence of moderate image noise and outliers. We conduct extensive simulations and real experiments to validate the performance of the new algorithm. The results demonstrate that RAS achieves notably higher accuracy than most existing robust motion-segmentation methods, including random sample consensus (RANSAC) and its variations. The implementation of the algorithm is also two to three times faster than the existing methods. The implementation of the algorithm and the benchmark scripts are available at http://perception.csl.illinois.edu/ras/.

Symmetry-based 3-D reconstruction from perspective images

Symmetry is an important geometric cue for 3-D reconstruction from perspective images. In this paper, we introduce a unified theoretical framework for extracting poses and structures of 2-D symmetric patterns in space from calibrated images. The framework uniformly encompasses all three fundamental types of symmetry, i.e., reflective, rotational, and translational, based on a systematic study of the homography groups in image induced by the symmetry groups in space. We claim that if a planar object admits rich enough symmetry, no 3-D geometric information is lost through perspective imaging. Based on two fundamental principles that utilize common spatial relations among symmetric objects, we have developed a prototype system which can automatically extract and segment multiple 2-D symmetric patterns present in a single perspective image. The result of such a segmentation is a hierarchy of new geometric primitives, called symmetry cells and complexes, whose 3-D structures and poses are fully recovered. Finally, we propose a new symmetry-based matching technique, which can effectively establish correspondences among the extracted symmetry cells across multiple images. We demonstrate the application of the proposed algorithms on image segmentation, matching, and 3-D reconstruction with extensive experimental results. The algorithms and systems are more accurate and easier to implement than existing point-or line-based methods.

Data-driven image completion by image patch subspaces

We develop a new method for image completion on images with large missing regions. We assume that similar patches form low dimensional clusters in the image space where each cluster can be approximated by a (degenerate) Gaussian. We use sparse representation for subspace detection and then compute the most probable completion. Our results show almost no blurring or blocking effects. In addition, both the texture and structure of the missing regions look realistic to the human eye.

Homography from Coplanar Ellipses with Application to Forensic Blood Splatter Reconstruction

Reconstruction of the point source of blood splatter in a crime scene is an important and difficult problem in forensic science. We study the problem of automatically reconstructing the 3-D location of the victim of a shooting from photographs of planar surfaces with blood splattered on them. We analyze this problem in terms of the multiple-view geometry of planar conic sections. Using projective invariants associated with pairs of conic sections, we match images of multiple conic sections taken from widely separated viewpoints. We further recover the homography between two views using the common tangents of pairs of conic sections. The location of the point source is then retrieved from the reconstructed scene geometry. We suggest how to extend these results to scenes containing multiple planar surfaces, and verify the proposed method with experiments on both synthetic and real images.

Segmentation of hybrid motions via hybrid quadratic surface analysis

In this paper, we investigate the mathematical problem underlying segmentation of hybrid motions: given a series of tracked feature correspondences between two (perspective) images, we seek to segment and estimate multiple motions, possibly of different types (e.g., affine, epipolar, and homography). In order to accomplish this task, we cast the problem into a more general mathematical framework of segmenting data samples drawn from a mixture of linear subspaces and quadratic surfaces. The result is a novel algorithm called hybrid quadratic surface analysis (HQSA). HQSA uses both the derivatives and Hessians of fitting polynomials for the data to separate linear data samples from quadratic data samples. These derivatives and Hessians also lead to important necessary conditions, based on the so-called mutual contraction subspace, to separate data samples on different quadratic surfaces. The algebraic solution we derive is non-iterative and numerically stable. It tolerates moderate noise and can be used in conjunction with outlier removal techniques. We show how to solve the hybrid motion segmentation problem using HQSA, and demonstrate its performance on simulated data with noise and on real perspective images.